Internet users increasingly rely on cloud-based services for social networking, shopping, gaming, and other activities. Some cloud-based services have achieved user bases of hundreds of millions of users and have extended their platforms to attract more users and/or keep existing users within the services' online ecosystems. For example, some cloud-based services have provided messaging mechanisms for users to privately and/or publicly exchange and/or share messages via these cloud-based services.
In order to encourage user participation, cloud-based services may maintain relatively low standards for user registration and submissions. Unfortunately, the large user bases available under these uniform platforms may attract those who wish to exploit large numbers of people, potentially bringing scams, phishing attempts, spam attacks, and/or unwanted content to users. Traditional classification systems for identifying and remediating illegitimate content may fail to correctly classify all content, resulting in undesired false positives and/or false negatives. Furthermore, traditional classification systems developed to combat email spam may rely on traditional email feature information, such as the mail servers of origin, email headers, user agents, etc. Accordingly, the instant disclosure identifies and addresses a need for additional and improved systems and methods for detecting illegitimate messages on social networking platforms.